RU2349990C2 - Method of electron-phonon drag - Google Patents

Abstract

FIELD: physics; electricity.

SUBSTANCE: electrodes composing rectifying contact with material are arranged on surface or in bulk of semiconductor material or material layer on semi-insulating or dielectric substrate. Furthermore electrode distance (D) is chosen considerably smaller than penetration depths into electric field material (L), (D<<L), caused by contact potential difference. Minimum electrode distance D_min=20 mcm, maximum electrode distance D_max=300 mcm. Before, after or during electrodes are to be arranged, or before, after or during electrode gap is made, the material is introduced with electronicvibrational centres (EVC) concentrated (N) within 2·10¹² cm^-3 to 3·10¹⁷ cm^-3. Material temperature is reduced to Debye temperature of, and temperature in material layer on substrate is reduced to Debye temperature of substrate phonons. Temperature difference ΔT=(T₂-T₁) is ensured between electrodes so that greater of temperatures T₁ and T₂ is not melting temperature of electrodes, material or substrate, and smaller of these temperatures is to be less than T_m-2θ, where θ is half-width of temperature dependence band of electron-phonon drag.

EFFECT: possibility to realise the effect at Debye temperatures of crystal phonons and possibility to control effect magnitude.

12 cl, 1 tbl, 11 dwg

Description

The technical field to which the invention relates.

The invention relates to the electronics of condensed materials, to methods for implementing and applying the thermoelectric effect of entrainment of electrons (holes) by phonons at Debye temperatures of phonons in solid materials and structures between electrodes. It can be used in microelectronics, radio engineering and electrical engineering as a principle of operation of solid-state electronic and optoelectronic devices with a reduced level of intrinsic noise without cooling. The invention implements a new physical mechanism for the drag of electrons by phonons at the Debye temperatures of phonons in solid-state materials and structures between electrodes. It is based on the use of intrinsic (Inherent, I-) vibrations and waves, the source of which is electron-vibrational centers (ECCs), namely, the interaction of I-vibrations and waves with electrons, holes and crystalline phonons in semiconductor materials and structures.

The theoretical foundations of modern solid-state electronics include the well-known adiabatic Born – Oppenheimer approximation [1], which is usually used to solve the Schrödinger equation for a crystal. In this approximation, the possibility of energy exchange between electrons and atomic nuclei in crystals is excluded. Obviously, the adiabatic Born-Oppenheimer principle limits the range of physical processes available for research and application in materials. Indeed, P. Dirac first showed [2] that this principle, generally speaking, is not fulfilled, and further studies [3] strengthened the understanding of the limitations of the adiabatic approach to the problem of solids in general and to solid-state electronics in particular. In this regard, the existing solid-state (semiconductor) electronics, which dominates in science and technology, can reasonably be called adiabatic electronics. This electronics does not answer many questions about the nature of crystals and physical phenomena in them, for example, such as superconductivity, hyperconductivity, superconductivity, which, in particular, is associated with the use of the adiabatic approach and limits the use of crystals.

On the other hand, the claimed invention is based on using the fundamental possibility of energy exchange between electrons and atomic nuclei in materials. Such electronics lies beyond the adiabatic principle (adiabatic approximation). It can reasonably be called non-adiabatic solid-state electronics. This non-adiabatic electronics includes the claimed invention and some well-known technical solutions in which violations of the adiabatic principle provide ECC. These centers in crystals create a channel of energy exchange between electrons and atomic nuclei, and technical solutions using such energy exchange represent a fundamentally new non-adiabatic solid-state electronics.

The prior art.

The effect of electron drag by phonons can manifest itself in the presence of a temperature gradient in the material in the form of an additional contribution to the thermopower (to the Seebeck effect). It differs from the well-known diffusion (drift) mechanism in that the drag of electrons by phonons occurs due to the coupling of electrons to the flow of phonons, due to the drag of electrons to the flow of phonons.

The phenomenon of differential thermoelectric power (Seebeck effect), discovered in 1821, is observed in the presence of a temperature gradient in materials and is characterized by the coefficient of thermoelectric power α = dE / dT, where E is the thermoelectric power and T is the temperature. This phenomenon was associated with the diffusion of mobile charge carriers, the concentration and mobility of which determine the value of this diffusion (drift) thermopower E _d . The coefficient of diffusion (drift) thermoelectric power (α _d ) in non-degenerate semiconductor materials is expressed by the following formula:

Where

n and p are the concentrations, μ _n and μ _p are the drift mobilities of electrons and holes, k is the Boltzmann constant, e is the elementary charge, r is the scattering parameter, F _n and F _p are the Fermi quasilevels of electrons and holes in a non-degenerate material [4, 5].

The value of α is equal to the entropy flux per unit electric current in the thermoelectric circuit [6].

It was established experimentally that the thermoelectric power in germanium single crystals increases with decreasing temperature from 200 K to 15 K [6–9], significantly exceeds the value calculated using formulas (1, 2) and to describe it, it is necessary to use an additional physical effect [10], which is called the effect of electron drag by phonons (UEF or Phonon drag).

In 1945, L. Gurevich first theoretically predicted the existence of such a new thermoelectric effect (E _f ), additional to diffusion thermoelectric power, caused by the drag of electrons by the phonon flux in metals [11] and noted the possibility of its existence in semiconductors. G.Picus in 1951 derived a formula for the coefficient of electrical energy in semiconductors:

where m ^* is the effective mass of the electron and ν is the speed of sound in the material, τ _f and τ _e are the relaxation times of phonons and electrons (broken brackets here indicate average values), but came to the conclusion that the effect is supposedly insignificant. The derivation of formula (3) can be found in [4, 12]. Thus, the thermoelectric power in the material is the sum of the diffusion thermoelectric power and the UEF effect: E = E _d + E _f , and the thermoelectric coefficient α = α _d + α _f .

A similar theory that takes into account the interaction between phonons and electrons in a material has been successfully applied to the analysis of experimental thermopower in doped Ge [13, 14]. It turned out that the contribution of the effect of electron drag by phonons to the thermopower is a band in which E _f reaches a maximum located between 15 K and 30 K. This band can be predicted from the data on the thermal conductivity of the material [14], but the thermal conductivity itself is almost independent of entrainment of electrons by phonons [11]. It was this band that was interpreted as the effect of electron drag by phonons (UEF), caused by the flow of only those phonons that are associated with electrons. The decrease in UEF corresponding to formula (3) with an increase in temperature above 30 K suggests that UEF exists only at low temperatures [15]. However, the thermoelectric power in bundles of carbon nanotubes in the range from 4.2 K to 300 K, measured in the laboratory of Nobel laureate R. E. Smalley, was presumably explained by the UEF effect [16], which is possible with strong electron-phonon coupling.

An effective coupling of electrons with phonons can be provided by electron-vibrational centers (ECCs) introduced into the material. Electronic transitions inside the ECC or external transitions to the energy levels of the ECC are associated with a significant number of phonons, and the equilibrium positions and vibration frequencies of the ECC depend on their electronic state and can change during electronic-vibrational transitions. That is why such centers and electronic transitions are called electron-vibrational centers and electron-vibrational transitions, respectively. A measure of the coupling of electrons with phonons at an ECC is the electron-phonon coupling constant S, which has the meaning of the average number of phonons involved in the electronic-vibrational transition. For ECC, it can reach 150, and the experimental S values in some samples exceed 22 [17-19], although in the absence of ECC the value S << 1 [20].

The probability of an electronic-vibrational transition depends on the number (p) of phonons absorbed or emitted [17-19]. If phonons of the same frequency are important, then the transition spectra consist of discrete lines corresponding to the participation of p = 0, 1, 2, ... phonons, and the energies of neighboring lines differ by the phonon energy. The highest probability of transitions corresponds to p≅S, which causes a (Stokes) shift of the spectrum maximum by S phonon energies relative to the phononless transition energy (corresponding to p = 0) to one of the energy levels of the ECC natural vibrations. If, in addition to phonons of one frequency, more than 3% of phonons of other frequencies participate in the electronic-vibrational transition, then the lines of the spectrum are broadened. If phonons of various types and frequencies participate in electronic-vibrational transitions, the spectrum turns into a continuous band located between the energies of the α, β, and γ types of intrinsic (Inherent, I-) vibrations, which are elastic vibrations of atomic nuclei (α -type), nuclei together with K-electrons (β-type) or nuclei of atoms together with K- and L-electrons (γ-type) relative to the rest of the electron shell or, which is the same thing, relative to the center of mass of the crystal [21 -23]. Natural vibrations in the atoms of the basic substance are possible in the presence of ECC. This ensures a strong electron-phonon interaction in the material.

The introduction of ECC into materials has a significant effect on their physical properties due to the strong electron-phonon coupling. For example, photoconductivity associated with ECC is negative. In the presence of an ECC, mobile electrons and holes are localized at the centers, their drift mobilities and the diffusion thermoelectric power described by formulas (1, 2) decrease or practically become zero, and the UEF effect due to electron-vibrational transitions between the ECC becomes possible to dominate.

, где ω_j - частота фононов типа j, участвующих в переходе, и согласно [11] выражается следующей формулой:The theory [17-19] describes both radiative and non-radiative electron-vibrational transitions (occurring without the participation of phonons), which are important for UEF, caused by elastic vibrations of the crystal with frequency ω. This theory is also applicable for transitions between ECCs, since in each material the centers are similar to each other and indistinguishable. The rate of nonradiative transitions of the center υ (ω) from state a to state b reaches a maximum at a frequency

, where ω _j is the frequency of phonons of type j participating in the transition, and according to [11] it is expressed by the following formula:

where υ _m is the maximum value of υ (ω),

(q_ja-q_jb) - изменение равновесных координат центра,

- среднее значение колебательного (квантового) числа. Вблизи максимума функции (4) величина (ω-ω_m) мала, степенной ряд в показателе экспоненты сходится быстро и можно ограничиться первым квадратичным членом ряда.

(q _ja -q _jb ) - change in the equilibrium coordinates of the center,

- the average value of the vibrational (quantum) number. Near the maximum of function (4), the quantity (ω-ω _m ) is small, the power series in the exponent converges quickly, and we can restrict ourselves to the first quadratic term of the series.

где

- постоянная Планка, ω_l - максимальная частота колебаний кристалла, индекс l обозначает тип колебаний. По этому правилу стали вводить температуры Дебая для акустических (А), оптических (О), продольных (L) и поперечных (Т) колебаний кристаллических решеток и для квантов таких колебаний - фононов [4, 5]. Температуры Дебая удобны для описания взаимодействия фононов с локальными (примесными) центрами и мы воспользуемся этими температурами. Определив по правилу Дебая температуру материала

, температуру материала при максимальной скорости переходов

, а также температуру

, полагая, что величина эффекта УЭФ пропорциональна скорости переходов υ(ω) и учитывая только первый член ряда в показателе экспоненты в (4), то можно записать температурную зависимость коэффициента α_ф в виде функции Гаусса:P. Debye introduced temperature in 1912

Where

is the Planck constant, ω _l is the maximum frequency of oscillations of the crystal, the index l denotes the type of oscillations. According to this rule, they began to introduce Debye temperatures for acoustic (A), optical (O), longitudinal (L) and transverse (T) vibrations of crystal lattices, and for quanta of such vibrations — phonons [4, 5]. Debye temperatures are convenient for describing the interaction of phonons with local (impurity) centers, and we will use these temperatures. Determining the material temperature according to the Debye rule

, material temperature at maximum transition rate

as well as temperature

, assuming that the magnitude of the UEF effect is proportional to the transition rate υ (ω) and taking into account only the first term of the series in the exponent in (4), we can write the temperature dependence of the coefficient α _f in the form of a Gaussian function:

where const is a temperature-independent quantity. The function α _f (T) reaches a maximum at a temperature T _m and is symmetrical with respect to T _m . If we take into account the first two terms of the series in (4), then α _f (T) can become asymmetric. We used function (5) to approximate the bands of the UEF effect in the temperature dependences of the thermopower of experimental samples of materials and structures.

To study the temperature dependences of the thermopower, we used samples of various materials and structures with a thickness of about 200 μm and an area of 14 × 8 mm ² containing ECC with S> 1. S values were determined from an analysis of the electron-vibrational IR absorption, reflection, or photoconductivity spectra of the samples. ECC was introduced into the test materials in various ways. So, samples of Si, InSb, InAs and layers of such materials with a thickness of 2 to 60 μm with concentrations of conduction electrons from 6 · 10 ¹² cm ^-3 to 1.68 · 10 ¹⁵ cm ^-3 on substrates of semi-insulating GaAs were irradiated with an electron flux of ~ 10 ¹⁸ cm ^-2 with an energy of 1 meV. According to the capacitance-voltage measurements, the ECC concentration in various samples ranged from 10 ¹⁴ cm ^-3 to 10 ¹⁶ cm ^-3 . Samples became partially compensated after irradiation with fast electrons, which is explained by the introduction of ECC into materials, which were formed mainly by impurity oxygen atoms. GaP samples with a concentration of aluminum or sulfur impurities of about 10 ¹⁵ cm ^–3 were subjected to heat treatment to activate ECC, which were formed by these impurity atoms. Highly oriented single crystals of pyrolytic graphite, epitaxial silicon layers 0.4 microns thick on sapphire substrates, and carbon nanotube films with a thickness of about 0.1 microns with a surface resistance of about 10 ³ Ohm / □ on quartz, fluorite, and yttrium aluminum garnet substrates were also investigated. (YAIG) made by sputtering graphite by an electron beam in a vacuum. According to the data on IR reflection spectra and activation energies of resistivity, ECCs in graphite and carbon nanotube films were formed by carbon atoms [21–23].

. Температуры измеряли с помощью термопар медь-константан с точностью 0,2 К. Температуру образца T=(T₂+T₁)/2 изменяли монотонно в течение 2-3 часов в криостате от 15 К, а в термостате - от 77 К до сотен градусов по шкале Кельвина. Величину термоэдс (Е) измеряли между контактами 1 и 2 с точностью 1 мкВ. Зависимость α(Т) измеряли "по точкам", вычисляя значение коэффициента термоэдс (УЭФ) при разных температурах образца, заменяя дифференциалы dE и dT конечными приращениями этих величин: α(T)=dE/dT=Е/(Т₂-Т₁). Зависимость α(Т) вблизи ее экстремумов аппроксимировали функцией (5).The temperature dependences of the thermopower were measured in the traditional way. The sample inclusion circuit is shown in the inset of FIG. Rectifying contacts to the samples were created using electrodes 1 and 2, which were applied to the surface of the material by thermal spraying of gold or aluminum in vacuum. In both cases, consistent measurement results were obtained. We placed the studied samples in a vacuum cryostat or in a thermostat with a nitrogen atmosphere at normal pressure. A temperature difference was created between the sample electrodes

. Temperatures were measured using copper-constantan thermocouples with an accuracy of 0.2 K. The sample temperature T = (T ₂ + T ₁ ) / 2 was monotonically changed for 2-3 hours in a cryostat from 15 K, and in a thermostat - from 77 K to hundreds of degrees Kelvin. The value of the thermoelectric power (E) was measured between contacts 1 and 2 with an accuracy of 1 μV. The α (T) dependence was measured “by points”, calculating the value of the thermoelectric coefficient (UEF) at different temperatures of the sample, replacing the differentials dE and dT with finite increments of these quantities: α (T) = dE / dT = E / (T ₂ -T ₁ ) The dependence α (T) near its extrema was approximated by function (5).

Figure 1 shows a typical temperature dependence of the thermoelectric power in silicon monocrystals containing ECC, which has relatively narrow bands with complex contours. It was noted that the extrema of the B, C, and D bands are located at the Debye temperatures of acoustic and optical phonons determined from data on silicon phonons [24]. Considering this fact and the polarities of the bands, bands B and C were explained by the effect of electron drag by acoustic ones, and band D by the effect of hole drag by optical phonons [25, 26].

 Typical temperature dependences of the thermoelectric power in InAs single crystals are presented in Fig. 2, and in GaP, in the inset of Fig. 2, on the left, and contain a number of bands [27]. The temperatures at the maxima of the bands in GaP and the bands G, H, I, K in InAs coincide with the Debye temperatures of various phonons. In individual InAs samples, a weak band is observed near a temperature of 424 K, which coincides with the sum of the Debye temperatures of two LA phonons. InSb samples, in addition to bands at the Debye temperatures of various LA and LO phonons, there are bands at temperatures of 134 K and 430 K, which are equal to the sum of the Debye temperatures of two phonons (2TA and 2TO).

Band A in Si (see Fig. 1), which has a maximum near 55 K, as well as the F band in InAs (see Fig. 2) with a maximum at 45 K, are most consistent with the interpretation as the UEF effect. However, in these defect-free materials there are no phonons with such Debye temperatures. Such phonons can arise as a result of the introduction of ECC into materials. Indeed, the dispersion branches of acoustic phonons under the influence of natural vibrations of atomic nuclei in a material change significantly [21, 25]. For a one-dimensional crystal model, these changes are qualitatively shown in the inset of Fig. 2, to the right. Coefficient δ describes the interaction between acoustic (A) and intrinsic (I) vibrations. For δ <0, a high density of phonon states with small values of the wave vector appears, and the Debye temperatures of these phonons can reach several tens of degrees. Such low-frequency phonons are apparently capable of creating experimentally observed low-temperature UEF bands (for example, bands A and F).

Figure 3 shows the characteristic temperature dependence of the thermopower of a single crystal of graphite, measured along the normal to the atomic planes of the graphite "parquet". It contains bands with maxima at the Debye temperatures of LO phonons and, apparently, at the Debye temperature of two (LO + LO) phonons whose energies are determined from the dispersion phonon branches of graphite [28].

In thin epitaxial layers of materials on substrates, thermoelectric bands are also observed, however, they are located at the Debye temperatures of the phonons of the substrate. The maximum temperatures of the M, N, and O bands in the InAs layer on the GaAs substrate (see Fig. 4) do not coincide with the Debye temperatures of phonons in InAs, but are located at the Debye temperatures of LA, LO, and TO phonons in GaAs. The inset of Fig. 4 shows the characteristic temperature dependence of the thermopower in the InSb layer on a GaAs substrate. This dependence has bands with maxima at the Debye temperatures of phonons in a GaAs substrate or at temperatures of 383 K, 440 K, 486 K, and 523 K, which are equal to the sum of the Debye temperatures of TO, LO, and LA phonon pairs in GaAs or pairs of similar phonons in InSb.

Fig. 5 and the inset of Fig. 5 show typical temperature dependences of the thermoelectric power in a silicon layer on a sapphire substrate and in a carbon nanotube film on a quartz substrate, having UEF bands whose maxima lie at the Debye temperature of the phonons of the substrates.

Narrow UEF bands are observed only in materials containing ECC. They are located at the Debye temperature of phonons; therefore, they are undoubtedly caused by electron-vibrational transitions between ECCs and definitely represent the UEF effect. We approximated the contours of the bands by dependence (5), using the values of T _m measured by us and choosing the values of θ and const.

The temperature T _m has the meaning of the Stokes shift of the maximum of the UEF band by the Debye temperature of one (p = 1) or by the sum of the Debye temperatures of several (p> 1) phonons relative to the phononless transition (p = 0, T _m = 0 K). According to the theory, the most intense band should be expected at a temperature coinciding with the sum of Debye temperatures p≅S phonons. However, in the experiments, temperatures T _m equal to the Debye temperatures of one (p = 1) or the sums of Debye temperatures of a pair of phonons (p = 2) are observed, although the electron – phonon interaction constant on the ECC in Si can reach S≥5, in InSb - S ≥3, in GaP - S≥4, in graphite and carbon nanotube films on substrates - S> 10. This result can be explained by the well-known decrease in S to values close to 1 due to the enhanced interaction between the ECCs when the centers approach each other as their concentration increases (above 10 ¹⁶ cm ^-3 ). Such a decrease in S explains the absence of UEF effect bands with the participation of three or more phonons (p> 2) at high ECC concentrations.

The temperature θ is the dispersion of the Gauss function (5). It has the meaning of the half-width of the UEF effect band at its half-height and represents the features of electron-vibrational transitions in a particular material. The most accurate approximation of the UEF bands by function (5) is achieved if the temperature θ has one specific value for all the UEF bands in the material. So, the value θ = 5.8 K was determined for the UEF bands in Si. It turned out that the band C in FIG. 1 is an overlap of three Gaussian bands with temperatures T _m corresponding to the Debye temperatures of various LA phonons. Each of bands B and D consists of two Gaussian bands with temperatures T _m corresponding to Debye temperatures of TA, LA, and LO phonons. The L band in graphite (see Fig. 3) consists of three Gaussian bands with temperatures T _m located at the Debye temperatures of various LO phonons with the value θ = 5 K. The bands in InAs and InSb have a half width θ = 4 K, in graphite θ = 5 K.

The UEF bands in InAs layers on GaAs substrates are characteristic of InAs θ = 4 K. The profile of the M band in the InAs layer (see Fig. 4) consists of two Gaussian curves with θ = 5 K. Each of the N and O bands is the sum bands with θ = 4 K and 25 K, θ = 4 K and 20K. The presence of the Gaussian components of some bands with temperatures θ> 10 K in narrow-gap materials, such as graphite, InAs, and InSb, can be presumably explained by the partial contribution of the diffusion thermoelectric power. Temperatures θ have the characteristic values for each material shown in Table 1.

Table 1
The temperatures θ in single crystals and single crystal layers of materials on semi-insulating or dielectric substrates Single crystals Material InAs Insb Si Ge Gap Graphite Cdhgte θ (K) 4.0 4.0 5.8 5,0 5,0 5,0 10.0 Layers of materials on substrates Material InAs Insb Si Carbon Nanotube Films Substrate Gaas Gaas Sapphire Quartz Fluorite Yalg θ (K) 4.0 4.0 4.7 5,0 5,0 2.6

Table 1 shows that the temperatures θ for single crystals coincide with the corresponding temperatures for layers of the same materials on substrates (with the exception of submicron Si layers on sapphire). The values of θ are the same for all UEF bands in each material [29] and reflect the properties of intracrystalline processes, which are independent of external conditions. Apparently, these are oscillatory processes of energy exchange between systems of electrons and atomic nuclei in a crystal that occur even in defect-free materials, which was shown theoretically using the example of a Si single crystal [30].

Thus, it has been established that in each ECC-containing material the temperature dependences of the thermoelectric power have narrow UEF bands with a Gaussian profile and the same width, located at the Debye temperature of the phonons of the material, and in the layers of the material on the substrate, at the Debye temperatures of the phonons of the substrate. These bands are caused by electron-vibrational transitions between the ECCs and represent the effect of electron drag by phonons. The results obtained are important because previously the UEF effect was observed only at low temperatures in the form of a band with a maximum between 15 K and 39 K and only in germanium single crystals. New data on the UEF effect in various materials at Debye phonon temperatures open up the possibility of applying this technical effect without cooling the materials.

Analogs of the claimed invention are known methods of implementation (creation) of thermoelectric power. In analogs of the invention, a semiconductor or metal solid-state material is usually used with electrodes forming electrical non-rectifying contacts to the material, create a temperature difference between the contacts of no more than 1 K ... 2 K, and measure the thermoelectric power arising in the material between the contacts [4, 5].

There is also a method disclosed in [13,14], in which a semiconductor Ge single crystal is used, electrodes are placed on its surface that form non-rectifying contacts to the material, create a temperature difference of ≈1 K between the electrodes, the material is heated from a temperature below 30 K and measure the temperature dependence of the thermoelectric power arising between the electrodes. This thermopower mainly represents the effect of electron drag by phonons (at temperatures no higher than 70 K).

Analogs of the invention do not allow the effect of electron entrainment by phonons at temperatures above 70 K, and even more so at Debye temperatures of phonons, which usually exceed 200 K, and in some materials they reach 800 K ... 1000 K.

As a prototype, you can specify the method disclosed in the publication: V. A. Vdovenkov, "The drag of electrons by phonons due to electronic-vibrational transitions in materials", Izvestiya VUZov, Materials of electronic technology, 2005, No. 1, pp. 65-70. In the known method, samples of semiconductor materials or layers of such materials with a thickness of up to 50 μm on semi-insulating or dielectric substrates with electrodes are used. Electron-vibrational centers (ECCs) were introduced into the samples. A temperature difference was created between the sample electrodes and the thermopower was measured. However, the known method does not disclose all the conditions necessary for the effect of electron entrainment by phonons at near-room and higher temperatures (essentially, at Debye temperatures of crystalline phonons).

SUMMARY OF THE INVENTION

The aim of this invention is to provide a method for implementing the technical effect of electron entrainment by phonons at near-room and higher temperatures (essentially, at the debye temperatures of crystalline phonons) in semiconductor solid-state materials and structures between electrodes, as well as methods for controlling the magnitude of this effect.

Such methods may be suitable for creating new solid-state electronic devices with a reduced level of intrinsic noise without cooling the devices.

An important task is the creation of electrical contacts to the material, which do not impede the existence and measurement of the thermoelectric power and allow a certain current to pass through the material or change the electric potential of the material in devices and devices.

Also important is the task of creating and controlling the temperature difference in the material by dragging electrons by phonons based on the use of intercenter electron-vibrational transitions.

According to paragraph 1 of the claims, in order to realize the effect of electron entrainment by phonons (UEF) in semiconductor materials or in layers of such materials up to 50 μm thick, on the semi-insulating or dielectric substrates, electrodes are placed on the surface or in the bulk of the material, forming rectifying contacts with the material, for example, Schottky contacts , choose the distance between the electrodes (D) is much less than the penetration depth of the electric field into the material (L), (D << L) caused by the contact potential difference, the minimum The distance between the electrodes is D _MIN = 20 μm, the maximum distance between the electrodes is D _MAX = 300 μm, before, after or during the placement of the electrodes or before, after or during the creation of a gap of width D between the electrodes, electron-vibration centers (ECCs) are introduced into the material ) in a concentration (N) from 2 · 10 ¹² cm ^-3 to 3 · 10 ¹⁷ cm ^-3 , bring the material temperature to the Debye temperature of the phonons of the material, and in the material layer on the substrate, it is brought to the Debye temperature of the phonons of the substrate, create a difference between the electrodes temperature ΔT = (T ₂ -T ₁ ), such that the largest of those temperature T ₁ and T ₂ did not reach the melting temperature of the electrodes, material or substrate, and the lower of these temperatures should be less than T _m -2θ, where θ is the half-width of the temperature dependence of the drag of electrons by phonons, measure the thermoelectric power between the electrodes, which under these conditions represents This is the effect of electron drag by phonons (UEF) in the material between the electrodes.

According to paragraph 2 of the claims, the electron-vibration centers are introduced only into the depletion region of the material or into the parts of the depletion region of the material adjacent to the electrodes between the electrodes.

According to paragraph 3 of the claims, in the method according to claims 1, 2, in order to determine the Debye temperatures and Debye frequencies of the crystalline phonons of the material, ΔТ <θ is selected, the temperature dependence of the UEF is measured, the temperatures of the maxima of this dependence (T _m ) are determined, and their Debye temperatures of crystalline phonons are equated , and the Debye frequency of the crystalline phonons of the material is calculated by the formula

According to paragraph 4 of the claims, in order to control the UEF value in the methods according to claim 2, the temperature difference of the electrodes (T ₂ -T ₁ ) is less than θ, (T ₂ -T ₁ ) <θ, the temperature of the material between the electrodes is changed at least within T _m ± 3θ.

According to paragraph 5 of the invention, in the methods according to claim 2, in order to control the magnitude of the UEF effect, the material or part of the material between the electrodes is illuminated in the absorption spectral region of the ECC or the material or in the spectral absorption region of the ECC and the material and the light density (I) is changed within from I = 0 to I = N _s / (ατ), where N _c is the effective number of electronic states in the conduction band, α is the optical absorption coefficient, and τ is the electron lifetime in the material between the electrodes.

According to paragraph 6 of the claims, in the methods of claim 2, in order to control the magnitude of the UEF effect, an additional field electrode or several field electrodes are used that form rectifying contacts to the material between the electrodes (pn junctions, metal-semiconductor or metal-dielectric contacts semiconductor), apply constant bias voltages between the material and various field electrodes or between different field electrodes, change the magnitudes of bias voltages and manipulate voltage polarities displacements.

According to paragraph 7 of the claims, in order to control the magnitude of the thermoelectric power in the method according to claim 2, an alternating voltage with an amplitude of less than half the height of the potential barrier to the electron in the contact, expressed in Volts, is applied between the material and one or more additional electrodes with a frequency acoustoelectric synchronism f = V / 2W, where V is the speed of sound in the material and W is the thickness of the material plate, or V is the speed of sound in the material layer on the substrate and W is the thickness of the layer on odlozhke or V - velocity of sound in the substrate and W - the total thickness of the layer and the substrate, changing the frequency of the AC voltage in the range of f / 2 to 2f or alter the amplitude of alternating voltage or change the frequency and amplitude of the alternating voltage.

,According to paragraph 8 of the claims, in the methods of claim 2, in order to control the UEF value and reduce internal noise in the material between the electrodes, a magnetic field transverse to the current lines is created, the intensity of which is selected in the range from 0 to

,

- постоянная Планка, р - номер электронно-колебательного уровня ЭКЦ, переходы на который используют для осуществления УЭФ (р=1, 2, 3, …, S).where m is the effective mass of the charge carrier (electron) in the material, k is the Boltzmann constant,

- Planck's constant, p - number of the electronic-vibrational level of the ECC, the transitions to which are used for the implementation of the UEF (p = 1, 2, 3, ..., S).

According to paragraph 9 of the claims, in the method of claim 8, the magnitude or direction or magnitude and direction of magnetic field induction is changed.

According to paragraph 10 of the claims, in the methods according to claims 1, 2, 4, 5, 6, 8, in order to reduce the level of internal noise of the UEF and increase the effect of the UEF in the material between the electrodes, a constant magnetic field is created with a strength of 0 to 2 Tesla, directed in parallel current lines in the material between the electrodes.

According to paragraph 11 of the claims, in the method according to claims 1, 2, 7, in order to control the UEF value, an alternating magnetic field is created in the material and its frequency is varied in the range from f / 2 to 2f, where f is the frequency of acoustoelectric synchronism, or its direction is changed induction, or change the frequency and direction of induction.

According to paragraph 12 of the claims, in the methods according to any one of claims 1, 2, 5, 6, 7, 8, 9, in order to obtain an additional to ΔT, the temperature difference between the electrodes creates along the streamlines in the material between the electrodes the concentration difference of the electron-vibration centers ( ΔN) or the difference in the concentrations of electrically active electron-vibrational centers, measure the UEF value, measure the temperature difference additional to ΔT.

The claimed invention is characterized by a combination of distinctive features, namely the introduction of a certain concentration of ECC into the material, the use of rectifying contacts to the material, the selection of permissible distances between the electrodes forming these contacts with the material, the placement of the electrodes on the material or in its volume, the choice of the working temperature of the material equal to the temperature Debye of one of the types of phonons of a material or substrate, the choice of the temperature difference of the electrodes, the ability to control the magnitude of the UEF effect from its elimination by illuminating the material or part of the material between the electrodes, as well as by applying a magnetic field of a certain magnitude and orientation and changing the magnitude of its induction, using additional field electrodes forming rectifying contacts with the material between the electrodes, and manipulating the bias voltages supplied to field electrodes relative to the material, by changing the temperature of the material, by applying a voltage between a material and electrodes with a certain amp cast and frequency.

Thus, the claimed method of implementing the effect of entrainment of electrons by phonons meets the criteria of the invention of "novelty."

Comparison of the claimed method for implementing the effect of entrainment of electrons by phonons with a prototype and other technical solutions in the art did not reveal technical solutions that have the specified set of distinctive features. This allows you to make a reasonable conclusion about the conformity of the claimed technical solution to the criteria of the invention "significant differences".

Really:

The claimed method allows the effect of electron entrainment by phonons in narrow temperature ranges near the Debye temperatures of phonons, which, in accordance with well-known data on Debye temperatures of various types of phonons in various materials, can range from 200 K to (800 ... 1000) K. Thus, the UEF effect can be carried out in materials near room temperature and at higher temperatures, which corresponds to the purpose of the invention.

To claim 1 of the claims.

The thickness of the layer on the substrate according to experimental data set no more than 50 microns. With a larger thickness of the material layer, a significant decrease in the UEF is observed, mainly due to the shunting of the rest of the material adjacent to the boundary with the substrate of the material lying outside this boundary layer, which is passive in the UEF. This is possible because the phonons of the substrate penetrate the material layer no further than several acoustic wavelengths, which corresponds to the attenuation expected from the calculations and the experimentally observed attenuation of the phonons of the substrate in the material layer. At layer thicknesses (≈50 μm), the UEF band attenuates at a Debye temperature of the substrate phonons, and with an increase in the layer thickness (more than 50 μm), the UEF band appears at a Debye temperature of the phonons of the material, and the contribution of the substrate phonons to the UEF becomes insignificant with increasing the thickness of the layer of material.

Contacts and distances between them.

An important condition is the design of the electrical contacts to the material and their position relative to each other. This can be seen by considering physical processes near the contacts. Figure 6 presents the energy band diagram of the structure with metal-semiconductor contacts (Schottky contacts) under conditions of thermodynamic equilibrium. Here F _m and F _PP denote the Fermi levels in the metal and semiconductor, respectively. The semiconductor has an electronic type of conductivity and F _pp close to the bottom of the conduction band E _c . E _v is the ceiling of the valence band, and the band gap of the semiconductor is E _g = E _c -E _v . The height of the built-in potential barrier eϕ _k is determined by the 2/3 rule, according to which the position of the Fermi level at the metal-semiconductor interface is 2/3 of the semiconductor band gap below the conduction band bottom, that is, 1/3 of the semiconductor band gap above the ceiling valence band [31]. From Fig. 6 it can be seen that the distance between the electrodes D is significantly greater than the penetration depth of the electric field into the semiconductor caused by the contact potential difference, i.e.

where ε is the relative dielectric constant of the semiconductor, ε ₀ is the electric constant, e is the electron charge, ϕ _k is the contact potential, n is the concentration of free charge carriers in the allowed energy band (electrons in the conduction band of the n-semiconductor or holes in the valence band p semiconductor) in the state of "flat zones". In this case, when measuring the thermoelectric power between the electrodes, the electrons will be scattered at the contact potential barriers. That is, the existing potential barriers of height eϕ _k prevent the measurement of the thermoelectric power in the material between the electrodes. This undesirable effect can be eliminated by choosing D significantly less than L. The corresponding zone energy diagram is depicted in Fig.7. 7 shows that when D is much less than L (D << L), the built-in potential barrier can be reduced to a negligible value. Moreover, the height of the barrier can be no more than kT. To do this, reduce D. At D much less than L (D << L), the semiconductor in the material between the electrodes has a hole type of conductivity and is separated from the rest of the semiconductor by a physical pn junction, and the built-in potential barriers are not able to scatter electrons, which can freely penetrate into metal, and the thermopower in the material between the electrodes can be detected by measuring the potential difference between the electrodes. In this case, the Schottky barrier prevents the penetration of electrons from the metal into the material between the electrodes and thereby excludes their influence on the UEF.

Similar energy diagrams can be considered in the case of a semiconductor with hole conductivity, assuming that the UEF in the material between the electrodes will arise as a result of hole transitions between the ECCs. However, the registration of the thermoelectric power between the electrodes is inevitably associated with the flow of current through the contact regions and, therefore, the process of recombination of holes at the interface with the metal is inevitable, which is equivalent to the process of hole scattering, which does not contribute to the increase in UEF. In addition, in this case, a physical pn junction does not occur and moving holes are able to enter from the semiconductor volume into the material between the electrodes and affect the UEF. However, hole-conduction material is in principle suitable for carrying out the present invention. Thus, the condition for the implementation of the UEF effect in the interelectrode gap is the condition of a significant smallness of D in comparison with L (D << L).

In addition, studies of the dependence of UEF on D have shown that UEF at maximum Debye temperatures reaches a maximum in all studied materials if D lies in the range from 20 μm to 300 μm, and outside these D values the UEF practically decays, which can be seen from Fig, which presents the relevant data for some materials. These data correspond to the relation between the UEF effect and the properties of the ECC and its weak dependence on the type of crystal lattice of the material and on the type of material.

ECC concentrations.

Obviously, for the implementation of UEF under the mechanism under consideration, which involves the participation of natural vibrations (and waves of such vibrations), at least the minimum concentration (N _min ) of electron-vibrational centers should be introduced into the material. These centers can equally well be introduced before, after, or during the deposition of electrodes on the material, or, which is the same, before, during, or after creating a gap between the electrodes of width D. The minimum concentration of centers N _min can be estimated based on that the interaction between the electron shells of the electron-vibrational centers is carried out by means of acoustic crystalline phonons, and such an interaction can be effective at a distance of the length of the corresponding acoustic wave H = U / F, where U is the velocity waves (speed of sound) and F is the frequency of the wave. In this case, N = H ^-3 . The minimum frequencies of phonons interacting effectively with ECC in crystals are close to 1.25 · 10 ¹⁰ s ^-1 . Based on the maximum speed of sound U = 9.79 · 10 ⁵ cm / s and the indicated frequency of the elastic wave for acoustic waves (in silicon), we obtain N _min = 2.6 · 10 ¹² cm ^-3 . The above estimate of N is valid for any crystal, since crystals have slightly differing lattice constants and the speed of sound in them. An analysis of the experimental photoconductivity spectra of silicon samples containing ECC (A-centers, which are associations of impurity oxygen atoms with silicon vacancies), allowed us to determine the minimum concentration of A-centers (from 2 · 10 ¹² cm ^-3 to 3 · 10 ¹² cm ^-3 ) capable of affecting the electrical properties of silicon. The maximum concentration of ECCs affecting the electrophysical properties of materials is close to 3 · 10 ¹⁷ cm ^-3 , which is confirmed by experimental data on the photoconductivity and tunneling of electrons through potential barriers during electronic-vibrational transitions.

The choice of ΔT = (T ₂ -T ₁ ).

Obviously, the largest allowable temperature difference ΔТ = (T ₂ -T ₁ ) for measuring the profile of the UEF effect band should be as small as possible and not exceed θ, since for large ΔТ it will exceed the half-width of the effect band at its half height, which is 2θ, the measured contour of the UEF band will be greatly distorted. These conditions for measuring the UEF are observed in the prototype. In our invention, there is another task, not just to measure the profile of the UEF band, but to measure the largest value of the UEF at the lowest internal noise. It is obvious that the value of the UEF is proportional to the integral of the contour of the UEF band over the temperature range. It will be maximum if the interval ΔТ covers the entire temperature region of the UEF band, mainly located between the temperatures T _m -2θ and T _m + 2θ. Therefore, the temperatures T ₁ and T ₂ must satisfy the conditions T ₁ ≥Т _m -2θ and Т ₂ ≤Т _m + 2θ ₂ , where Т _m is the temperature at the maximum of the band, equal to the Debye temperature of the phonon, and θ is the half-width of the band at its half height. Figure 9 shows such a situation. Figa is a graph of the temperature of the material between the electrodes on the distance between them. The temperatures of the electrodes are T ₁ and T _2, respectively, the Debye temperature is T _m . The contour of the corresponding UEF band is shown in Fig. 9b under the indicated measurement conditions: T ₁ ≥T _m -2θ and T ₂ ≤T _m + 2θ with a maximum at T = T _m . It can be seen from FIGS. 9a and 9b that only a region of material with a thickness ΔX contributes to the UEF. It can be seen that if the temperature range covers temperatures T _m -2θ and T _m + 2θ and exceeds it, then the integral of the strip contour will tend to a constant value, and with temperature changes over a wider range, it will remain constant. Therefore, temperature fluctuations in this case will not cause a change in the UEF signal, which ensures a decrease in the level of internal (thermal) noise when measuring the UEF.

Moreover, in our case, thermal noise can exist due to temperature fluctuations only in the band approximately (T _m -2θ, T _m + 2θ). Therefore, the temperature band (thermal noise) in our case does not exceed 3θ ... 4θ, which is several times less than the width of the traditional temperature band of thermal noise, equal to T _m . Figure 10 shows the expected (calculated) dependence of the UEF value on the width of the temperature interval ΔТ measured in units of θ, and the points of the corresponding experimental dependence for GaAs at D = 50 μm are shown. Figure 10 shows that the UEF value reaches “saturation” and is practically independent of temperature at ΔT> 4θ, which corresponds to reduced (thermal) noise and the purpose of the invention. In this case, the temperatures T ₁ and T ₂ should not reach a different Debye temperature, as well as the melting temperature of the material or substrate, and the lowest of these temperatures should be less than T _m -2θ.

It is important to note that the method of measuring UEF according to claim 1 of the claims can be used to control the homogeneity of the thermoelectric properties of the material between the electrodes. Indeed, in this case, the temperatures T ₁ and T _{2 are} changed in such a way that the layer of thickness ΔX in which the UEF arises moves from one electrode to another (see Fig. 9). At certain temperatures T ₁ and T _2, it is easy to determine the position of the layer ΔX. Therefore, scanning the ΔX layer and measuring the UEF, we can determine the dependence of the UEF on the coordinate of the layer. In the inhomogeneous material, in the dependence shown in FIG. 10, in the “plateau” region, deviations from the UEF “saturation” value appear, reflecting the heterogeneity of the thermoelectric properties of the material (caused, for example, by the heterogeneity of the distribution of the ECC concentration in the material between the electrodes). In the case of a homogeneous material, deviations from the value of “saturation” of the UEF will not be observed. It is important to note that an increase in the temperature gradient (T ₂ -T ₁ ) / D helps to increase the resolution of the method for controlling the uniformity of the material, because contributes to a decrease in ΔX.

If ΔT reaches and exceeds the difference in the Debye temperatures of phonons, then, depending on the UEF versus ΔТ, as ΔТ increases, two (or more) UEF growth regions and two (three or more) "plateaus" corresponding to the contributions of two (or more) UEF bands.

To claim 2 of the formula. The claimed method can be carried out in cases where the ECC is introduced only in the part of the material between the electrodes adjacent to the electrodes, since the interaction between the ECC located in these parts is carried out by exchanging phonons. This feature is proposed to be used in claim 2 of the claims.

To claim 3 of the formula. It has been proposed to use the established correspondence of the temperatures of the maxima of the narrow UEF bands in ECC materials to the Debye temperatures of crystalline phonons to determine the temperatures and Debye frequencies of phonons of materials. These can be monolithic semiconductor materials or layers of materials on substrates, as well as semiconductor and dielectric substrates. Thus, by affordable, simple and inexpensive thermoelectric method, measuring the temperature dependence of UEF, it is possible to determine the temperature and Debye frequencies in semiconductor and dielectric materials, which is important for the study of new materials and structures, without the use of complex and expensive experimental methods, such as commonly used for such targets the scattering of slow neutrons. This simple possibility of determining the Debye parameters of the vibrational spectrum of crystal lattices of various materials is proposed for use in claim 14 of the claims.

To claim 4 of the formula. If the temperature difference of the electrodes (T ₂ -T ₁ ) <θ, then the deviation of the temperature of the material between the electrodes from the temperature of the maximum UEF (T _m ), which coincides with the Debye temperature of the phonons that cause this thermoelectric effect, by up to 3θ and more is accompanied by a significant decrease in UEF , which can be seen from the temperature dependences of the UEF shown in Fig.1-5. In paragraph 4 of the claims it is proposed to use this temperature dependence to control the magnitude of the UEF.

To claim 5 of the formula. Illumination of the material between the electrodes or a part of the material between the electrodes in the spectral absorption region of the ECC or the material creates additional electronic and electronic-vibrational transitions that affect the value of the UEF, which is proposed to be used in claim 5 of the claims. The highest density (intensity) of illumination can be determined by assuming that illumination does not substantially change the position of the Fermi level (quasi-Fermi level for electrons) in a material. According to the theory of recombination of electrons and holes in semiconductors [4, 5], this corresponds to the concentration of electrons in the conduction band of the material n no more than the effective density of electronic states in the conduction band

- эффективная масса для плотности состояний,

, n=αIτ≤N_с, где α (см^-1) - коэффициент оптического поглощения материала, I (фотонов·см^-2·с^-1) - плотность освещения, τ(с) - время жизни электронов. Поэтому допустимую плотность освещения можно определить из следующего соотношения I≤N_с/ατ.Where

is the effective mass for the density of states,

, n = αIτ≤N _s , where α (cm ^-1 ) is the optical absorption coefficient of the material, I (photons · cm ^-2 · s ^-1 ) is the light density, τ (s) is the electron lifetime. Therefore, the allowable density of illumination can be determined from the following ratio I≤N _s / ατ.

It should be noted that the photoconductivity often used for detecting radiation, in our case associated with the ECC, is negative, i.e., the conductivity of the material between the electrodes decreases with increasing intensity of the detected radiation, while in the known photodetectors the photoconductivity is positive. In this case, however, the UEF increases with increasing intensity of the recorded illumination, since almost all charge carriers (both electrons and holes, as established experimentally) are localized at the ECC and participate in electron-vibrational intercenter transitions, increasing the UEF. In addition, the temperature range of the UEF is not more than 4θ << Т _m , θ is constant for each material, which ensures a lower level of internal noise of the UEF compared to the noise level of known photodetectors [32, 33, 34], in which the noise is related to the temperature band lying between 0 K and the working temperature of the material T K, the width of which coincides with the working temperature of the material. Therefore, in the claimed method, the noise value is less, since it is determined by the value of θ, which is less than the working temperature of the material T = T _m and does not depend on it.

The method according to claim 5 of the claims can be used to register optical radiation, which provides a reduced level of intrinsic (internal) noise of a photodetector operating in the vicinity of the Debye temperature of phonons. The desired operating temperature of such a photodetector can be set by selecting and applying a semiconductor (substrate) with the corresponding Debye temperature of phonons, and the spectral sensitivity region can be selected by choosing a semiconductor with the corresponding band gap. The spectral sensitivity region of such a photodetector is composed of the intrinsic (fundamental, fundamental) absorption region of the applied semiconductor and the electron-vibrational absorption bands (ECC) located at optical quantum energies less than the band gap, which corresponds to the theory of electron-vibrational transitions and is confirmed experimentally.

It is advisable to choose the illumination intensity in the range from I = 0 to I = N _s / (ατ), where N _s is the effective number of states in the allowed energy zone of the material, α is the optical absorption coefficient of the material, and τ is the lifetime. Indeed, at a higher light intensity, the concentration of carriers in the material is so high (αIτ >> N _s ) that changing it no longer affects the electronic-vibrational transitions and the UEF value.

To claim 6 of the formula. The use of an additional field electrode or several field electrodes forming rectifying metal-semiconductor or metal-dielectric-semiconductor contacts to the material between the electrodes, the application between the material and various field electrodes, generally speaking, different bias voltages and manipulation of the magnitudes and polarities of these voltages allows you to change the length and the shape of the streamline between the electrodes, which is equivalent to a change in the distance between the electrodes and causes a change in the value of the UEF. This ability to control the UEF is used in paragraph 6 of the claims.

The use of two (or more) additional field electrodes allows for a wider (for example, in comparison with a traditional field-effect transistor) range of changes in the value of the UEF depending on the voltages on the field electrodes.

Obviously, the magnitude of the bias voltage of reverse polarity on any of the additional electrodes should not reach the voltage of its breakdown, and the magnitude of the direct bias should not reach the height of the potential barrier in the contact (ϕ _to ), measured in Volts.

The method according to claim 6 of the claims can be used to create an analog of a field effect transistor with the difference that instead of the difference in electrical potentials between the source and drain of the field effect transistor, the invention applies a temperature difference between the electrodes. In addition, using the method according to item 12 of the formula, a similar device can be implemented without the difference in temperature of the electrodes, but by creating an inhomogeneous distribution of concentrations (electrically active ECC). A skepticism towards these methods (according to paragraphs 6 and 12) is not excluded due to the relatively low UEF values, usually ranging from units to hundreds of millivolts, however, for example, living brain cells work at about the same voltages, provide an enviable reliability, stability, low energy consumption throughout the life of organisms. True, this vital activity of organisms is preserved in a narrow temperature range (2θ wide) near the characteristic Debye temperatures of organic semiconductor structures, cells, and molecules. Such a relatively narrow temperature range of the life of organisms is a kind of payment for high reliability, durability and low level of intrinsic noise, as well as for high noise immunity in relation to the action of external electric, magnetic and electromagnetic fields.

To claim 7 of the formula. In modern manufacturing technology of semiconductor devices, semiconductor wafers with a thickness of W = 200 ... 300 μm are usually used. Devices are formed on one (front, working) flat surface, and the other on the back surface, the crystal is soldered to the holder (heat sink). Such plates of materials with plane-parallel surfaces are for phonons (for sound in a crystal) a resonator similar to the Fabry-Perot optical resonator. For this sound (phonon) resonator, a certain frequency of acoustoelectric synchronism is characteristic f = 1 / t, where t is the time of flight of the sound wave (sound) from the front surface to the back surface of the plate and back with the speed of sound V. Flight time t = 2W / V. The corresponding frequency of acoustoelectric synchronism is f = V / 2W. If a voltage with the frequency of acoustoelectric synchronism f is applied between the electrode and the semiconductor, then due to the modulation by this voltage of the height of the potential contact barrier in the presence of an ECC and an increased phonon density in the contact region, the speed of the electron-vibrational transitions changes and, accordingly, the UEF value changes. This property of acoustoelectric synchronism in paragraph 5 of the claims is proposed to be used to control the magnitude of the UEF. In layered structures, other frequencies of acoustoelectric synchronism are also important, such as when V is the speed of sound in the substrate and W is the total thickness of the layer and substrate, and also when V is the speed of sound in the layer and W is the thickness of the layer on the substrate. These types of acoustoelectric synchronism are practically implemented and are important, since they affect the electronic-vibrational transitions and the value of the UEF.

The amplitude of the acoustoelectric synchronism voltage should not reach the approximate value ϕ _k / 2, where ϕ _k is the height of the potential barrier in the contact, expressed in Volts. Otherwise, injection currents occur that can disrupt acoustoelectric synchronism. In practice, it turned out that the voltage of acoustoelectric synchronism with an amplitude of 1 to 500 microvolts is sufficient.

To claim 8 of the formula. A magnetic field. It is possible to estimate the magnetic field induction, if we start from the frequency of the acoustic phonon 1.25 · 10 ¹⁰ s ^-1 participating in the electronic-vibrational transitions of the ECC. Indeed, the ECC can be represented as a charged oscillator with mass m and electron charge e. The equation of motion of the oscillator, taking into account the action of the Lorentz force in a magnetic field, can be written as follows:

, учитывает действие силы Лоренца, когда скорость и индукция В взаимно ортогональны. Данное уравнение допускает осциллирующее решение при условииwhere X is the generalized (configurational) coordinate, ω is the cyclic oscillation frequency, B is the projection of the magnetic field induction on the normal direction to the velocity of the charge carrier. Speed term

, takes into account the action of the Lorentz force when the speed and induction B are mutually orthogonal. This equation allows an oscillating solution provided

In other words, oscillatory movements of an oscillator with frequency ω are possible when B is not too large. If the oscillation frequency of the electron associated with the ECC is fixed and determined by the properties of the electron-vibrational center, then with an increase in B to a certain value determined by the indicated inequality, the center oscillations will become impossible. If an electron participates in a complex center vibration with several independent frequencies, then, as B increases, the oscillations with frequencies in sequence in order of increase will subsequently be suppressed by the magnetic field. In principle, it is possible to choose a value of B at which oscillations of electron-vibrational centers with any frequency will be suppressed. This seems to be the mechanism for suppressing ECC oscillations in a magnetic field. Putting m equal to the effective mass of the electron and ω = 2π · 1.25 · 10 ¹⁰ s ^-1 , from inequality (6) we obtain the minimum value of the critical magnetic field strength for silicon ≈0.25 T. The critical induction of the magnetic field will take multiples of the minimum critical value, which corresponds to the participation of p = 1, 2, 3, ..., S or more phonons in the UEF effect.

Figure 11 shows a typical dependence of the thermoelectric power at the Debye temperature of the phonon LA in a silicon single crystal containing A centers (≈10 ¹⁴ cm ^-3 ) on the magnetic field induction, which shows the minima located at the calculated values of B. In addition, from 11 shows that magnetic fields with an induction of not more than 1.8 T have an effect on the UEF, which corresponds to the participation of approximately (S +2) phonons in the UEF. Considering the mechanism of interaction with phonons and a magnetic field in semiconductors that is common to all ECCs and inequality (7), we can determine the largest induction necessary to suppress UEF in any material.

подавляет участие фононов с циклической частотой

в однофононных электронно-колебательных переходах и вызывает уменьшение УЭФ вблизи соответствующих значений индукции магнитного поля. Переходы ЭКЦ с участием одного фонона с частотой ω возможны на электронно-колебательные уровни энергии

, где номер уровня р=1, 2, 3, …, подавляются магнитным полем с индукцией

. Кроме того, на эти же электронно-колебательные уровни ЭКЦ, вообще говоря, возможны переходы электронов с участием 2-х, 3-х, 4-х, … фононов и связанные с ними вклады в УЭФ. Из фиг.11 видно, что минимумы зависимости УЭФ от индукции В почти достигают нуля при вычисленных значениях В. Следовательно, вклад в УЭФ 2-х, 3-х, 4-х, …, фононных процессов является незначительным. Следовательно, для подавления вклада в УЭФ электронно-колебательных переходов с участием фононов частоты ω на p-й энергетический уровень необходима индукция магнитного поля

. Выбирая индукцию между 0 и

можно изменить величину УЭФ, уменьшить шум УЭФ от переходов на уровни с меньшими значениями p, что предусмотрено п.8 формулы изобретения.According to formula (6), the direct magnetic field directed along the normal to the stream line in the material between the electrodes with an induction of at least

suppresses the participation of phonons with cyclic frequency

in single-phonon electronic-vibrational transitions and causes a decrease in the UEF near the corresponding values of the magnetic field induction. ECC transitions involving a single phonon with frequency ω are possible at electron-vibrational energy levels

where the level number p = 1, 2, 3, ..., are suppressed by a magnetic field with induction

. In addition, generally speaking, electron transitions involving 2, 3, 4, ... phonons and related contributions to UEF are possible to these electronic vibrational levels of the ECC. From Fig. 11 it can be seen that the minima of the dependence of the UEF on induction B almost reach zero at the calculated values of B. Therefore, the contribution to the UEF of 2, 3, 4, ..., phonon processes is insignificant. Therefore, in order to suppress the contribution of electron-vibrational transitions with the participation of phonons of frequency ω to the pth energy level in the UEF, the induction of a magnetic field

. Choosing an induction between 0 and

you can change the UEF value, reduce the UEF noise from transitions to levels with lower p values, which is provided for in paragraph 8 of the claims.

Эта оценка B_max верна и в других (кроме кремния) полупроводниках, так как значения S в них мало различаются, а при сравнительно высоких концентрациях ЭКЦ в них S близко к 1. Поэтому максимальное значение индукции В, необходимое для полного подавления колебаний ЭКЦ и соответственно УЭФ, можно принять раной 2 Тл.In addition, in Fig. 11, the minima of the dependence of the UEF on В with the number p = S + 2 are visible, where S = 5 is the electron – phonon coupling constant at the A centers in silicon. It is seen that the electron-vibrational levels up to p = 7 are active. Therefore, to suppress electron oscillations at the ECC, magnetic field induction is sufficient

This estimate of B _{max is} also valid in other (except silicon) semiconductors, since the S values in them differ little, and at relatively high ECC concentrations in them, S is close to 1. Therefore, the maximum value of induction B necessary to completely suppress ECC vibrations and, accordingly, UEF, you can take a wound 2 T.

To claim 9 of the formula. The dependence of the speed of electronic-vibrational transitions and, accordingly, the value of the UEF on the magnitude and direction of the magnetic field relative to the direction of the streamlines in the material between the electrodes, which is determined by the action of the Lorentz force on moving charges, is proposed for use to control the value of the UEF in paragraph 9 of the claims.

To claim 10 of the formula. A magnetic field with an induction of from 0 to 2 T, directed along streamlines in the material between the electrodes, suppresses the electron motion oscillating with the phonon frequency during electron-vibrational transitions normal to the current lines. This leaves electrons with only one degree of freedom of movement along streamlines. According to the well-known theorem of thermodynamics on the equilibrium distribution of kinetic energy over degrees of freedom, an electron has only kT / 2 energy (instead of 3kT / 2 in the absence of a magnetic field), where k is the Boltzmann constant, which corresponds to much smaller thermodynamic fluctuations, therefore, lower noise of electron vibrational transitions, and, consequently, lower UEF noise, and also contributes to an increase in UEF. This possibility of reducing internal noise and increasing UEF is proposed for use in clause 10. claims

To claim 11 of the formula. The dependence of the UEF on the magnitude and direction of an alternating magnetic field with frequencies from f / 2 to 2f in the material between the electrodes is proposed to be used to control the UEF value in claim 11 of the claims. Such a regulation of the UEF value is possible, since a magnetic field (by means of the Lorentz force) with a frequency in the indicated vicinity of the frequency f influences the acoustoelectric synchronism and UEF due to the fact that the acoustic waves excited in this case stimulate the acoustoelectric synchronism and electronic-vibrational transitions depending on magnitude, frequency and direction of the magnetic field relative to streamlines in the material between the electrodes.

To item 12 of the formula. In the presence of a difference in the concentrations of electron-vibrational centers along streamlines in the material between the electrodes, additional intercenter electron-vibrational transitions arise, causing an additional contribution to the effect of electron drag by phonons and to the temperature difference between the electrodes. The necessary difference in ECC concentrations can be created with the introduction of centers. You can also create the corresponding difference in the concentrations of electrically active ECCs, for example, by injecting electrons into the part of the material between the electrodes from the outside or irregularly illuminating the material between the electrodes.

The invention is illustrated by drawings.

Figure 1 shows the temperature dependence of the thermopower in Si. The inset on the left shows a diagram of the inclusion of a sample of material when measuring the thermopower (E). T ₁ and T ₂ are material temperatures near electrodes 1 and 2. The inset on the right shows dashed Gaussian curves 3, 4, 5, the sum of which, shown by a solid curve, describes the contour of strip C.

In figure 2. The temperature dependence of the thermopower in InAs is shown. Dotted lines show Gaussian curves approximating the brightest bands. The inset on the left shows the temperature dependence of the thermopower in doped gray GaP. The inset on the right shows the dispersion curve of the natural vibrations of nuclei in atoms (I) and the dispersion curves of acoustic vibrations for δ <0, δ = 0, δ> 0, where δ is a parameter that describes the relationship between the natural vibrations of atomic nuclei and acoustic vibrations.

In figure 3. The temperature dependence of the thermopower in pyrolytic graphite, measured along the normal to the atomic planes, is shown. The inset shows the approximation of the strip L by a solid curve, which is the sum of three dashed Gaussian curves (6, 7, 8).

In figure 4. A typical temperature dependence of the thermoelectric power in an InAs layer on a semi-insulating GaAs substrate is shown. Curve 9 was measured when the sample was heated, curve 10 — when it was cooled. The inset shows a similar dependence of the thermoelectric power in the InSb layer on the GaAs substrate. Dotted lines show approximating Gaussian curves.

Figure 5 shows the characteristic temperature dependence of the thermopower in a silicon layer on a sapphire substrate. The inset shows the temperature dependence of the thermopower in a carbon nanotube film on a quartz substrate. Dotted lines show approximating Gaussian curves.

Figure 6 shows the energy band diagram of the metal-semiconductor-metal structure at a distance between the contacts D (along the X direction normal to the metal-semiconductor interface, along the electric field line in the semiconductor) exceeding the electric field penetration depth (L) caused by contact potential difference in the material (D >> L).

Figure 7 shows the energy band diagram of the metal-semiconductor-metal structure with a distance between the contacts D is much less than the depth of penetration of the electric field L into the material (D << L).

Fig. 8 shows typical dependences of the thermoelectric power measured between Schottky contacts in SiC, Ge, InAs, GaAs, GaP, InP materials containing ECC (~ 5 · 10 ¹⁵ cm ^-3 ) and representing the UEF effect at the Debye temperature of LA phonons, depending from the distance between the contacts D << L.

On figa dependence 11 represents the temperature distribution in the material between the electrodes. On figb shows the profile of the UEF strip with a maximum at a temperature of T _m . Dashed lines show how the width of the UEF band determines the thickness of the layer ΔX in which the UEF effect occurs. Dependence 12 represents a different temperature distribution in the material between the electrodes (at different values of T ₁ and T ₂ ), corresponding to a different position of the ΔX layer in which the UEF effect occurs.

In figure 10, the light squares indicate the expected (calculated) dependence of the UEF value on the temperature difference between the electrodes, measured in units of θ: ΔT / θ = (T ₂ -T ₁ ) / θ, from which it is clear that for ΔT> 4θ there is " plateau ”, i.e. “saturation” of the UEF value is achieved. Dark circles show the experimental results for the UEF effect caused by LA phonons in GaAs at D = 50 μm.

Figure 11 shows a typical dependence of the thermoelectric power, representing the UEF effect in silicon, at the Debye temperature of the phonons LA (D = 50 μm << L), on the induction of the magnetic field (B) directed normal to the current lines in the material between the electrodes.

Using the claimed invention allows to improve the properties of known semiconductor devices (for example, photodetectors) and to ensure the operation of fundamentally new electronic devices with reduced intrinsic (internal) noise due to the fact that they are active only in narrow temperature ranges of width (3 ... 4θ) near T _m . The intrinsic noise of devices based on the UEF effect and operating at a Debye temperature (usually exceeding 200 K) is determined not by the operating temperature T _m , but by the width of the temperature interval ≅4θ << T _m and turn out to be much smaller than in known devices. Indeed, the most typical types of noise, such as radiation (photon), temperature, Johnson, generation and recombination, depend on the width of the temperature interval between 0 K and the operating temperature T K, presented in the noise characteristics as a function of T [32, 33, 34 ]. To reduce noise, they usually resort to cooling devices, i.e. to narrow the temperature range (0 ... T). Therefore, the use of the UEF effect as the principle of operation of the devices ensures the achievement of reduced internal noise compared to the noise of known devices by narrowing the noise temperature interval to a width of about 4θ << T _m , that is, several times compared to the operating temperature T _m . Such a new technical situation is possible due to the fact that the electron-vibrational transitions between the ECCs causing the UEF effect are tightly coupled with the energy exchange between the atomic nuclei and the electron system in crystals. This oscillating energy exchange is characterized by the temperature θ, is determined by the nature of the atoms of the crystal, and is independent of the temperature of the crystal. That is why the noise temperature of the UEF effect 4θ << Т _m and, accordingly, the noise of devices operating on the basis of this effect is significantly lower than the noise of drift and diffusion currents used in known devices. This provides new devices based on the use of the UEF effect at Debye temperatures of phonons with significant noise and functional advantages.

Moreover, another method is known for reducing internal noise by reducing the number of degrees of freedom for charge carriers (electrons and holes). This principle is realized in semiconductor quantum-dimensional structures (Raman structures), which are also called semiconductors with superlattices [35, 36], where the motion of charge carriers is possible only in thin two-dimensional layers of matter and carriers have not three, but only two degrees of freedom. In accordance with the well-known theorem of thermodynamics on the equal distribution of kinetic energy over the degrees of freedom of particles forming a thermodynamic system at a temperature T, the energy kT / 2 is on average for each degree of freedom. In cattle, each charge carrier has only two degrees of freedom (two-dimensional electron gas, 2D gas) and has a lower average energy equal to 2 · (kT / 2) = kT. Therefore, the energy fluctuations and the corresponding internal current noise in the Raman spectroscopy are much lower than in bulk single crystals. The use of structures with one-dimensional electron motion (with one degree of freedom of charge carriers), the so-called quantum filaments, allows for even lower internal noise of devices.

The claimed invention uses inter-center electron-vibrational transitions, which are one-dimensional in nature and, therefore, can reduce internal noise to a lower level than in cattle, and even lower than in devices based on quantum fibers. Experimental studies of the activation energies of the resistivity of semiconductors containing electron-vibrational centers, as well as their optical absorption spectra (spectra of electron-vibrational transitions) were carried out [22]. Studies have shown that the energy levels of the ECC are described by the formula of a linear (one-dimensional) harmonic oscillator: E (ν) = E ₀ (ν + 1/2), where ν is the vibrational quantum number taking the values 0, 1, 2, .... In the experiments, the levels determined by "zero-point" ECC corresponding to ν = 0 and equal to E _0/2, where E ₀ - quantum own (I-) EKC oscillations. In the case of two-dimensional and three-dimensional oscillations, the “zero” energies would be equal to 2 · E _0/2 = E ₀ and 3 · E _0/2 . Thus, it was shown that ECCs are one-dimensional oscillators having one degree of freedom. In this regard, the electron – vibrational transitions between the ECCs are also one-dimensional (directed along the temperature gradient, which follows from the theory of such transitions [11, 17-19]). This corresponds to the least noise. Thus, it turns out that ECC-containing semiconductors provide a reduced noise level in devices based on the drag of electrons by phonons in comparison with devices based on cattle or quantum fibers. So, the internal noise of infrared photodetectors based on ECS-containing materials (Ge, Si, InSb, InP, InAs, GaAs) turned out to be less than the noise of Raman-based devices and on quantum fibers. If we take into account the very complex and expensive manufacturing technology of cattle and quantum filaments, as well as devices based on them, in comparison with devices based on bulk semiconductors or semiconductor layers containing ECC on substrates, then the advantage of the claimed invention seems to be undeniable.

Let us evaluate the frequency properties of devices using electronic-vibrational transitions. For evaluation, we assume that the device operates in the UEF band caused by acoustic phonons, the energy of which, for example, in silicon, is approximately 22 meV (T _m about 265 K), which corresponds to the frequency of acoustic phonons f = 5 · 10 ¹² s ^-1 . In each electronic-vibrational transition, an average of S phonons is involved. The probability of the participation of S phonons in one process is less than the probability of the participation of one phonon in approximately S! time. Therefore, the expected operating frequency can reach f / S !. For silicon, S≤5. Therefore, the operating frequency of the device can reach 42 GHz. If the device operates in the UEF band caused by optical phonons, then the operating frequency can reach 85 GHz. Considering that in a material with a high ECC content (more than 10 ¹⁵ cm ^-3 ) the value of the constant S decreases to 1, the operating frequencies can reach hundreds of GHz at low levels of intrinsic noise, with a simple and cheap technology for manufacturing devices based on the use of the UEF effect in materials containing electron-vibrational centers. Thus, the speed of devices using electronic-vibrational transitions of the ECC is, in principle, limited not by the flight time of charge carriers, as in known devices, but by the frequencies of phonons and the degree of their connection with centers, which provides the devices with improved characteristics.

Electronic-vibrational centers are introduced into materials by various known methods (using heat treatment, alloying, radiation treatment). Radiation irradiation of materials is one of the most technologically advanced ways of introducing ECC into semiconductor materials and structures. In this regard, during the operation of devices using ECC and electronic-vibrational transitions, under the conditions of radiation exposure, ECC will be introduced into the material, which will not impair the operability of such devices in a wide range of radiation doses. Therefore, devices based on the use of electronic-vibrational transitions, in principle, have increased radiation resistance in comparison with known devices.

INFORMATION SOURCES

1. Born M., Oppenheimer. Ann. D. Phys. 1927. V.84, No. 4. P.457.

2. Dirac P.A.M. Proc. Cambridge Phil. Soc. 1930. V. 26. P. 376.

3. Doltsinis N. L., Marx D. Journ. of theor. and Camput. Chem. 2002. V.1, No. 2. P.319-349.

4. Anselm A.I. Introduction to the theory of semiconductors. - M .: Nauka, 1978. - S.557-564.

5. Seeger K. Physics of semiconductors. - M.: Mir, 1977 .-- P.113.

6. Herring C. // Phys. Rev. 1941. V. 59. P.889. - Same: Herring S., Nichols M.H. // Revs. Modern Phys. 1949. V.21. P.185.

7. Lark-Horovitz K., Middleton A.E., Miller P., Scanlon W.W. and Wallerstein J. // Phys. Rev. 1946. V.69. P.259.

8. Frederikse H.P.R. // Phys. Rev. 1952. V.86. P.647. - Same: Phys. Rev. 1953. V. 91. P. 491. - Same: Phys. Rev. 1953. V. 92. P.248.

9. Johnson V.A., Lark-Horowitz K. // Bull. Am. Phys. Soc. 1953. V.28, No. 2. P.32.

10. Geballe H.H. // Phys. Rev. 1953. V. 92, No. 3. P.857.

11. Gurevich L. // Journ. Of phys. 1945. V.9. P.477; 1946. V.10. P.67. - The same .: JETP. 1946.T.16. S.193-228.

12. Stlbans L.S. // Physics of semiconductors. - M .: Owls. Radio, 1967. - S.299-304.

13. Herring C. // Phys. Rev. 1953. V. 92. P.857. - The same: Phys. Rev. 1954. V. 96. P.1163.

14. Geballe T.N. and Hull G.W. // Phys. Rev. 1954. V. 94. P.1134.

15. Tsidilkovsky I.M., Matveev V.A. // In the book: Physical Encyclopedia. - M .: Owls. Encyclopedia, 1990. - Vol. 2. S.76-77.

16. Hone J., Ellwood I., Muto M., Mizel A., Cohen M.L., Zettl A., Rinzler A.G., Smally R.E. // Phys. Rev. Lett. 1998. V.80, No. 5, p. 1042.

17. Pekar S.I. Research on the electronic theory of crystals. - M., Gostekhizdat, 1951.204 p.

18. Huang K. and Rhys A. // Proc. Roy. Soc. 1950. V.A204. P.406.

19. Pekar S.I. // JETP. 1952.V.22, issue 6, p.641. - The same .: Physics – Uspekhi. 1953.Vol. L, issue 2, p .97.

20. Toyazawa, T. // Techn. Report of ISSP (Tokyo). 1961. Ser. A, No. 16.

21. Vdovenkov V.A., Kosakovskaya Z.Ya., Kolosov V.V., Mokerov V.G. // DAN. 1999. T.365, No. 5, S.611-613.

22. Wodenkov V.A. // Izv. Universities. Materials electron. technicians. 2003. No1. S.57-62.

23. Vdovenkov V.A. // arXiv: cond-mat \ 9904299; arXiv: cond-mat \ 0207215.

24. Stierwalt D.L. and Potter R.F. // Phys. Rev. 1965. V.137. S.A1007.

25. Vdovenkov V.A. // arXiv: cond-mat \ 0207218. - Same: High technology. 2002. No4. S.55-60. - The same: Microsystem technology. 2002. No. 12. S.17-22.

26. Vdovenkov V.A. // Materials of the Intern. scientific and practical conf. Intermatik-2003 “Fundamental Problems of Radioelectronic Instrumentation”. M., 2003. P.69-70.

27. Vdovenkov V.A. // Materials of the Intern. scientific and technical. conf. "Thin films and layered structures." M., 2002, part 2. S.33-40. - The same .: Materials of the Intern. practical and practical. conf. Polymaterials-2003 "Interphase relaxation in polymaterials". M., 2003. Part 2. S.111-112. 25.

28. Smith G., Wakabayashi N. Dynamic properties of solids and liquids. - M .: Mir, 1970. S.144-146.

29. Vdovenkov V.A. Temperature bandwidth of the effect of electron drag by phonons in materials. Materials of the International Scientific and Practical Conf. INTERMATIC-2004, “Fundamental Problems of Radioelectronic Instrument-Making”, pp. 102-105, Moscow, 2004. The same: V. A. Vdovenkov. Entrainment of electrons by phonons due to electronic-vibrational transitions in materials. University News Materials of electronic technology, No. 1, 2005, p.65-70.

30. Pastore G., Smargiassi E., Buda F. // Phys. Rev. A. 1991. V.44. No. 10. P.6334-6347.

31. Zee. C. Physics of semiconductor devices. Moscow, Mir, 1984, v. 1, p. 283.

32. Markov M.N. Infrared receivers. M., "Science", 1968, pp. 40-48.

33. Anisimova I.D., Vikulin I.M., Zaitov F.A., Kurmashev Sh.D. (edited by V.I.Stafeev). Semiconductor photodetectors: Ultraviolet, visible and near infrared spectral ranges. - M .: Radio and communications, 1984. - 214 p.

34. Van der Zil A. Noises during measurements. M .: Mir, 1979, p. 69-112, 129-220.

35. M.Herman. Semiconductor superlattices. M .: Mir, 1989, 238 p.

36. Base F., Bulgakov AA, Tetervov AP. High-frequency properties of semiconductors with superlattices. M .: Nauka, 1989, 287 p.

Claims (12)

1. A method for implementing the effect of electron entrainment by phonons (UEF) in semiconductor materials or in layers of such materials with a thickness of up to 50 μm on semi-insulating or dielectric substrates, including the placement of electrodes, the introduction of electron-vibration centers (ECCs), creating a temperature difference between the electrodes, characterized in that electrodes forming rectifying contacts with the material, for example, Schottky contacts, are placed on the surface or in the bulk of the material, while the distance between the electrodes (D) of the the depth of penetration of the electric field (L), (D << L) caused by the contact potential difference, the minimum distance between the electrodes D _MIN = 20 μm, the maximum distance between the electrodes D _MAX = 300 μm, before, after or during placing electrodes either before, after or during the creation of a gap between electrodes of width D, electron-vibrational centers (ECCs) are introduced into the material in a concentration (N) of 2 · 10 ¹² cm ^-3 to 3 · 10 ¹⁷ cm ^-3 , the temperature of the material is adjusted to the Debye temperature of the phonons of the material, and in the material layer on the substrate, to Debye temperature of the substrate phonons, between the electrodes create a temperature difference ΔT = (T ₂ -T ₁ ), such that the largest of the temperatures T ₁ and T ₂ does not reach the melting temperature of the electrodes, material or substrate, and the smaller of these temperatures should be less than T _m -2θ, where θ is the half-width of the temperature dependence of the drag of electrons by phonons. 2. A method for realizing the effect of electron entrainment by phonons according to claim 1, characterized in that the electron-vibration centers are introduced only into the depletion region of the material or into the parts of the depletion region of the material adjacent to the electrodes between the electrodes. 3. The method of implementing the effect of electron entrainment by phonons according to any one of claims 1, 2, characterized in that they create a temperature difference between the electrodes of less than θ, ΔT = (T ₂ -T ₁ ) <θ, measure the temperature dependence of the UEF, determine the temperature of the maxima of this dependences (T _m ) corresponding to the Debye temperatures of crystalline phonons, and the cyclic Debye frequencies of the crystalline phonons of the material are calculated by the formula

4. The method of implementing the effect of electron entrainment by phonons according to claim 2, characterized in that the temperature difference of the electrodes ΔT = (T ₂ -T ₁ ) is less than θ, (T ₂ -T ₁ ) <θ, the material temperature between the electrodes is changed within T _m ± 3θ. 5. The method of implementing the effect of electron entrainment by phonons according to claim 2, characterized in that the material or part of the material between the electrodes is illuminated in the absorption spectral region of the ECC or material or in the spectral absorption region of the ECC and the material and the illumination density (I) is varied within I = 0 to I = N _s / (ατ), where N _c is the effective number of electronic states in the conduction band, α is the optical absorption coefficient, and τ is the electron lifetime in the material between the electrodes. 6. The method of implementing the effect of electron entrainment by phonons according to claim 2, characterized in that they use an additional field electrode or several field electrodes, apply constant bias voltages between the material and various field electrodes or between different field electrodes, change the bias voltages and manipulate the voltage polarities displacement. 7. The method for implementing the effect of electron entrainment by phonons according to claim 2, characterized in that an additional field electrode or several field electrodes are used, an alternating electric voltage with an amplitude less than half the height of the potential barrier to the electron in the contact, expressed in contact, is applied between the material and one or more electrodes Volts, and with the frequency of acoustoelectric synchronism f = V / 2W, where V is the speed of sound in the material and W is the thickness of the material plate, or V is the speed of sound in the material layer on the substrate W - the thickness of the layer on the substrate, or V - velocity of sound in the substrate and W - the total thickness of the layer and the substrate, changing the frequency of the AC voltage in the range of f / 2 to 2f or alter the amplitude of alternating voltage or change the frequency and amplitude of the alternating voltage. 8. The method of implementing the effect of entrainment of electrons by phonons according to claim 2, characterized in that they create a magnetic field transverse with respect to the current lines in the material between the electrodes, the intensity of which varies from 0 to

where m is the effective mass of the charge carrier (electron) in the material, k is the Boltzmann constant,

- Planck's constant, p - number of the electronic-vibrational level of the ECC, the transitions to which are used for the implementation of the UEF. 9. A method for realizing the effect of electron drag by phonons according to claim 8, characterized in that the magnitude or direction or magnitude and direction of magnetic field induction is changed. 10. A method for realizing the effect of electron entrainment by phonons according to any one of claims 1, 2, 4, 5, 6, and 8, characterized in that a constant magnetic field with induction from 0 to 2 Tesla directed parallel to the streamlines is created in the material between the electrodes in the material between the electrodes. 11. A method for realizing the effect of electron entrainment by phonons according to claims 1, 2 and 7, characterized in that an alternating magnetic field is created in the material and its frequency is varied from f / 2 to 2f, where f is the frequency of acoustoelectric synchronism, or the direction of its induction or change both the frequency and direction of induction. 12. A method for realizing the effect of electron entrainment by phonons according to any one of claims 1, 2, 5, 6, 7, 8, and 9, characterized in that along the streamlines in the material between the electrodes create a difference in the concentration of electron-vibrational centers (ΔN) or the difference in the concentrations of electrically active electron-vibrational centers, measure the UEF value, measure the temperature difference additional to ΔT.

Images